We are looking for talented individuals to join our team in 2027. As a graduate, you will get opportunities to pursue bold ideas, tackle complex challenges, and unlock limitless growth. Launch your career where inspiration is infinite at our Company.
Successful candidates must be able to commit to an onboarding date by end of year 2027. Please state your availability and graduation date clearly in your resume.

Team Introduction: Join the Graphics Engine Team , where we push the boundaries of digital interaction by developing proprietary physics simulation, animation, and rendering systems.
Our team merges pioneering research with robust engineering to build sophisticated frameworks and mathematical models that power dynamic simulations, advanced animation, and rendering solutions, while prioritizing real-time performance. We leverage modern machine learning technologies, to emulate complex physics interactions at speeds multiple orders of magnitude faster than traditional methods, and to tackle challenging computer graphics problems that no conventional technique can solve. We are committed to pioneering new frontiers in technology, ensuring remains at the forefront of innovation in the industry.

Top content: Graphics rendering serves as core infrastructure for multimedia applications. Amid the rapid growth of short-video and interactive entertainment, the traditional production pipeline—marked by long cycles, high costs, and steep technical barriers—has severely limited creative expression for mainstream users. In recent years, breakthroughs in AIGC (such as emerging representations like NeRF and 3DGS) have driven 3D generation from experimental research toward commercial deployment, establishing it as a critical foundation for future spatial intelligence.This topic focuses on cutting-edge 3D generation technologies, aiming to evolve the 3D production pipeline from the traditional "modeling-simulation-rendering" workflow to a new AIGC-driven paradigm: "AI Objects – AI Scenes – AI Animation – AI Rendering." Through this transformation, the project seeks to drastically lower barriers to 3D creation, fully empower the interactive entertainment ecosystem, and build a robust technical foundation for spatial intelligence.
Topic Challenges:
-Insufficient 3D data: High-quality native 3D model data remains severely scarce globally (fewer than 10 million assets), lagging far behind image, text, and video data. This creates a critical data bottleneck for 3D generation. To meet massive generation demands, we must develop more efficient data filtering and collection pipelines, and solve the challenge of extracting high-quality 3D and interactive data from massive short videos and livestreams at low cost.
-Extremely high barriers and complexity in 3D foundation model training: 3D generative models are true foundation models with extremely high training barriers, requiring deep expertise in distributed multi-GPU training and extensive iterative experimentation. In practice, we must resolve complex engineering challenges including long-sequence memory management and compute resource allocation, while achieving optimal algorithmic balance between photorealistic quality, multi-view 3D consistency, and generation controllability.
-Lack of physical interaction and editing capabilities in AI Rendering: Despite rapid advances in AI rendering (e.g., 3DGS), real-world deployment is limited by compute power, visual quality, and consistency. For instance, Gaussian splatting struggles with lighting interaction and physical simulation. Moreover, without explicit geometry like traditional 3D meshes, AI rendering exhibits clear limitations in handling complex surfaces (e.g., hair), supporting fine-grained editing, and maintaining physical consistency.
-Effective integration of advanced video generation into 3D rendering: Video generation models excel in dynamic capture and realism but often lack 3D spatial consistency and physical controllability. A core future challenge is to clarify the interplay between video generation and 3D engine rendering, effectively transfer powerful spatial-temporal semantic priors from video models into 3D rendering, compensate for implicit generation limitations, and transition from passive video generation to active interactive 3D scene creation.

Qualifications

Minimum Qualifications:
-Individuals who are completing or recently completed a PhD in Software Development, Computer Science, Computer Engineering, or a related technical discipline.

Preferred Qualifications:

-Modeling experience in one or more of the following areas: Ads, Search engine, Recommender System, NLP/CV.
-Have a solid foundation in algorithms related to LLMs, including but not limited to comprehensive learning and practical experience in areas such as single-modal/multi-modal LLM application and deployment.
-Strong publications record in top conferences or journals（e.g., ICLR, NeurIPS, ICML, ACL, EMNLP, NAACL, CVPR, ICCV and ECCV）